Soil Temperature Prior to Veraison Alters Grapevine Carbon Partitioning, Xylem Sap Hormones, and Fruit Set

Biomass partitioning and nonstructural carbohydrates

Altering soil temperature from budbreak was previously shown to have little effect on the total dry biomass of grapevines at anthesis, but temperature differences caused dry biomass to be partitioned differently (Field et al. 2009). Soil warming between budbreak and anthesis promoted shoot biomass accumulation and leaf area development at the expense of root starch reserves (Field et al. 2009, Rogiers et al. 2011). Conversely, when a cooler soil temperature regime was maintained across the same period, shoot biomass was reduced, but the rate of mobilization or utilization of root starch was also decreased. The effects of soil temperature regime in the period between anthesis and veraison were consistent with these vine responses. That is, there was no overall difference in whole vine biomass, but a significantly lower root:shoot ratio with warmer soil. Soil warming caused a decline in root starch concentrations of previously cooled vines, while starch concentrations of the continuously warmed plants remained low. Under cooled soil conditions, all vines had significantly higher starch in the roots at veraison. This was particularly pronounced for the previously warmed vines, which more than doubled the amount of stored starch from anthesis to veraison. These results illustrate that soil temperature can exert considerable influence on the balance between starch accumulation (storage) and utilization (annual growth). The enhanced aboveground growth of grapevines in sites where the soil warms rapidly in spring has been noted previously, particularly in cool sites (Jackson 2001). It is now clear that this response is attributable to accelerated utilization of carbohydrate reserves.

Another interesting finding was that fruit biomass at the time of veraison was significantly higher in vines that had previously been cooled prior to anthesis. This is possibly attributed to these plants having significantly greater root reserves at anthesis to support early berry growth. Many studies have reported that mobilization of carbohydrate reserves are used to help negate the effects of reduced assimilation (i.e., through leaf defoliation) on fruit development (see review by Holzapfel et al. 2010).

Flowering and fruit set

In grapevines, the transition from flower to berry relies on photoassimilates, normally from basal leaves (Keller 2015). Concurrent demand for those assimilates from growing shoots and restoration of carbohydrate reserves limit their availability for fruit set, as inferred from treatments that enhance fruit set, such as trunk girdling or shoot tipping, or those that diminish it, e.g., severe pruning, defoliation, or leaf shading (see review by Holzapfel et al. 2010). Our study revealed that soil temperature-induced tensions in carbohydrate demand during flowering influence the degree of fruit set. Lower fruit-set percentage was induced by cooling previously warmed roots. Tabing et al. (2013) also found that reducing the root temperature from 20 to 10°C for a two-week period over flowering reduced fruit set in cv. Chardonnay vines. Interestingly, in this study, the marked reduction in fruit set was associated with the increase in root carbohydrates. Poor fruit set has been attributed to competition with rapid shoot growth (May 2004). However, after removing potential effects of different flower numbers at anthesis (by regression analysis), plants with the lowest shoot growth rate (elongation or dry biomass) between anthesis and veraison had significantly less fruit set than all other treatments. This is attributable to the fact that restoration of root carbohydrate reserves was also occurring during this period to a greater extent (~200% greater) than all other treatments. This demonstrates a strong acclimation that favors reserve accumulation over maximal fruit set, at least when demand by both functions is simultaneous. It also suggests that viticultural practices such as shoot tipping and application of growth retardants to improve fruit set are not likely to be effective when root carbohydrate reserves are low and restoration is favored by cool soil temperature.

Fruit set remained unaffected when the cool soil temperature was maintained postanthesis. This can be largely attributed to their higher reserve status at the onset of flowering. Continuously warmed plants also had unimpaired fruit set. Accelerated foliar development, including a high proportion of fully autotrophic proximal leaves possibly accounted for this. Rogiers et al. (2014) also showed that constant soil temperature up until fruit set (warm, ambient, or cool) had no effect on the percentage of fruit set. Cool/warm-treated vines, with greater mobilization of reserves to the shoot during flowering to veraison, interestingly had no effect on fruit set. Biomass analysis suggests these reserves were utilized solely in shoot growth.

Poor fruit set is sometimes attributed to restricted nutrient uptake from cold soil. However, no significant differences in N content (g/vine) between treatments and the normal fruit set in the continuously cooled plants indicate that nutrient uptake was not a factor in this regard. Furthermore, improved nutrient uptake seen in vines grown in a warmer soil had no effect on the percentage of fruit set in a similar experimental setup (Rogiers et al. 2014, Clarke et al. 2015).

CKs and ABA

transZR was the major CK form in the xylem sap at the times of flowering and fruit set and is consistent with the CK profile of xylem sap in many other plants (Emery and Atkins 2002). However, nucleotide forms that were not detected at flowering (Field et al. 2009) were apparent both at fruit set and veraison, when DHZRP was one of the main dominant CK forms along with transZR. Thus, marked changes in the xylem sap CK profile occurred across the season. Notably, at the time of veraison, transZR concentrations were considerably higher in vines subjected to cool soil temperature postanthesis.

Soil temperature treatment had little or no effect on stomatal conductance, photoassimilation rate, or transpiration of the most recently fully expanded leaves at anthesis (Field et al. 2009). However, by the completion of fruit set 40 days later, leaf gas exchange measurements of similar aged leaves were significantly lower in vines grown in cooler soil. It is well known that low soil temperatures decrease transpiration by reducing absorption of water directly by decreasing the permeability and hydraulic conductivity of roots to water and indirectly by increasing the viscosity of water (Kozlowski 1987). Those effects may have contributed to both the lower transpiration rate and the lower midmorning leaf water potential of the cooled plants. As leaf area increased markedly between the time of flowering and veraison in the present study for vines in all treatments, the total canopy transpiration would have increased concomitantly. If low root conductance had occurred in cooled vines, then supply of water to the vines would have been impeded in this treatment. Leaf transpiration rates in cooled vines were indeed significantly reduced compared to vines in warmer soil, which is consistent with a restricted supply of water. Reductions in stomatal conductance apparently mediated the low transpiration rates of the cooled vines to maintain the vine water potential.

ABA has been shown in numerous studies to reduce transpiration by closing stomata in response to water stress (Keller 2015). However, our study did not convey this with warmed vines, with assumed higher stomatal conductance, having higher ABA concentrations in the xylem sap at veraison. Veselova et al. (2005) found that differences in stomatal opening caused by different root temperatures were not caused by ABA. Furthermore, their work revealed that decreased levels of CK in the xylem sap closed stomata of wheat seedlings when roots were cooled. CKs are known to be involved in regulating stomatal conductance, with application of synthetic CK generally opening the stomata and reversing the effect of ABA on stomata under water stress (Stoll et al. 2000). However, application of benzylaminopurine (BAP) at low concentrations to leaves of sugar maple (Reeves et al. 2007) reduced stomatal conductance by up to 40%. Furthermore, when ZR and iPA were introduced into the transpiration stream of well watered Arubutus unedo plants, there was no stimulatory effect on transpiration, with ZR even reducing transpiration compared to control plants (Burschka 1985, as cited in Fuβeder et al. 1992). Therefore, the higher concentrations of riboside CKs in the xylem sap of vines exposed to cool soil conditions at veraison relative to those exposed to warm conditions could possibly be a root-to-shoot signal that closed the stomata in response to low temperature-induced reductions in root hydroconductivity.

Leaf CK concentrations did not differ greatly nor was there a consistent pattern between the soil treatments. However, the CK profile in leaf tissue differed from the xylem sap, with DHZR, ZROG, transZRP, and iPRP being the predominant CK forms at both sample times. These differences in CK profiles between xylem sap and leaf tissue highlight the different and changing roles of CKs from fruit set to veraison. The dominance of transZ CK types, particularly transZR relative to iP types in the xylem sap, highlight the biased distribution of these CKs within the plant. Root-derived transZ CKs are typically found in the xylem, and Ip-type CKs are typically found in the phloem (Hirose et al. 2008). This compartmentalization of CKs suggests selective transport systems for transZ and iP types and their role in acropetal and systemic long distance signals (Hirose et al. 2008). Grafting experiments using multiple isopentenyltransferase (a key gene in CK biosynthesis) Arabidopsis thaliana mutants highlighted the importance of root-derived transZ CKs and their transport from root to shoot being necessary in shoot development (Matsumoto-Kitano et al. 2008). In the current study, no cisZ CKs or O-glucosides were detected in the xylem sap, further supporting the role of CK compartmentalization and dominance of CK transport in the xylem and not CK processing or local synthesis during this time. At veraison, xylem sap transZR was reduced relative to fruit set; this may reflect a switch from CK long distance signaling from root-shoot CKs in the xylem toward a preference for local CK production in the leaves and fruit. This switch in CK dominance would reflect the developmental changes that occur postveraison, i.e., fruit ripening. Böttcher et al. (2015) noted a decreased contribution of transZ CKs and an increased concentration of iP in grape berries during fruit ripening in four grapevine cultivars. Böttcher et al. (2015) also identified 38 CK-related genes and examined their expression across 16 weeks post-flowering in developing Shiraz grape berries. Developmental changes in the expression of genes related to CK biosynthesis, processing, etc., suggested that local CK biosynthesis more likely contributes to the postveraison accumulation of iP within the examined grapevine berries (Böttcher et al. 2015).

Long distance transport as well as local CK production in the Shiraz grapevine leaves during fruit set and veraison is reflected in the abundance of CK forms detected. Nucleotide CKs are precursors to active CKs and are considered the first product in CK biosynthesis; these are modified to semiactive ribosides that are further modified to the active freebase forms (Sakakibara 2006). This sequence of CK processing can be modified through direct activation of nucleotide forms to freebases (further explanation Frébort et al. 2011). The detection of higher levels of nucleotides, particularly iPRP (at veraison) in the leaves indicate local CK production may become more important during this development change, as well as reflect the potential shift from transZ to iP CK type dominance.

The shifts in CK concentration within xylem sap as well as leaf tissue may reflect the changing role of xylem-derived root CKs toward local CK production. Thus, xylem sap CK involved in the mobilization of starch-derived solutes early in the season may possibly change roles to regulating leaf stomatal conductance later in the season when leaves are able to supply their CK needs.